1. Field of the Invention
This invention relates to a novel core glass composition that is particularly suited for, but not limited to, optical waveguiding signal amplifier articles due, in part, to the flatness of its gain spectrum; and to a non-CVD method for making continuous clad filaments. More particularly, the invention relates to a novel rare earth doped, fluorinated aluminosilicate glass composition and to a cullet-in-tube method for making continuous clad filament such as, e.g., optical waveguiding fiber and conductive conduit.
2. Description of Related Art
This application is related to provisional U.S. application Ser. No. 60/034,472, filed on Jan. 2, 1997, which is herein incorporated by reference in its entirety.
Erbium doped optical amplifiers and, in particular, erbium doped fiber amplifiers, have revolutionized optical telecommunications by providing all optical high gain, low noise amplification without the need for costly repeaters. Current erbium doped fiber amplifiers, however, are not well suited for multi-channel amplification due to the variation of their gain spectrum as a function of wavelength, denoting gain flatness or the lack thereof. As used herein, the term "gain flatness" will refer to the change in the shape of the gain spectrum over a particular wavelength range; i.e., flat gain means substantially equal gain for all wavelengths over the wavelength range of interest. For erbium, the wavelength range of interest is from about 1530 nm to 1560 nm. When the gain spectrum is not sufficiently flat over this wavelength interval, multiple channels at different wavelengths are not uniformly amplified, making high data rate communication systems inoperable, impractical and unaffordable.
The art teaches that co-doping an erbium doped fiber with Al.sub.2 O.sub.3 increases Er solubility and results in a flatter gain spectrum than that exhibited by an erbium doped pure silica fiber. However, currently known erbium doped aluminosilicate compositions yield a best performance gain flatness of about 27 dB of gain deviation per 100 dB of gain over a 32 nm wide band, and are prone to devitrification at high levels, i.e., greater than a few weight percent of Al.sub.2 O.sub.3. Compositions of SiO.sub.2 doped with La.sub.2 O.sub.3 or La.sub.2 O.sub.3 +Al.sub.2 O.sub.3 have also shown improved gain flatness and rare earth solubility, but the Al.sub.2 O.sub.3 and La.sub.2 O.sub.3 concentrations are all well below 1%, and thus have but a minor effect.
Fluoride glasses, e.g., ZBLAN (57ZrF.sub.4 --20BaF.sub.2 --4LaF.sub.3 --3AlF.sub.3 20NaF (mole %)) exhibit good gain flatness and low phonon energy, but these compositions require pumping at 1480 nm due to up-conversion effects, and exhibit increased noise over 980 nm pumping. Moreover, they are difficult to fiberize, are not fusion spliceable to silica fibers, are prone to devitrification, and have poor durability.
In addition to gain flatness, the gain provided by a rare earth doped host medium is also a parameter of interest. Higher gains can theoretically be achieved by increasing the concentration of the suitable rare earth dopant; however, above a modest concentration, rare earth ion clustering and concentration quenching become a problem. Current CVD methods for glass fiber preform fabrication are quite limited by the composition ranges of the host glasses. Only modest amounts of rare earth elements can be incorporated without clustering effects, and volatile components such as alkalis and halogens cannot be introduced because of their tendency to vaporize during laydown. Furthermore, other important glass modifiers, e.g., alkaline earths, cannot be incorporated due to lack of high vapor pressure CVD precursors. Even if a glass soot can be deposited by CVD, it must subsequently be consolidated which can lead to crystallization or loss of glass components with high vapor pressures.
The inventors have therefore recognized a need for a glass composition and optical waveguide articles made therefrom that accommodate high levels of rare earth doping without clustering to provide high gain; that provide an improvement in gain flatness over conventional compositions; that are pumpable at 980 nm for good noise performance; that are fusion spliceable to conventional silica based fibers; that match or exceed the durability of conventional optical fiber waveguides; and that are easy to fabricate. Related application Ser. No. 60/034,472 describes a novel glass-ceramic composition and devices made therefrom that exhibits many of the advantageous requirements described above. The inventors have discovered, however, that the composition of the instant invention provides those advantages while eliminating the additional ceraming step required to make the glass-ceramic, and further provides a wider constituent composition range.
Accordingly, there is a need for a method for making waveguiding optical fiber from a wide variety of glass and glass-ceramic compositions, and other continuous clad filament articles such as, e.g., conductive conduit, that overcomes the disadvantages of the known methods, and that is more practical, efficient, and economical than conventional methods. The "cullet-in-tube" method of the invention allows almost any glass that can be produced by chemical (sol-gel, vapor deposition, etc.) or physical (batch and melt) techniques, or other feedstock in granular or powder form ("cullet" as referred to herein), to be economically fabricated in the form of a continuous clad filament. The rapid quenching of this technique allows for unstable glasses and glass-ceramics to be fiberized.